Archive for June 2012
Hi, first thanks for the awesome bots! I've noticed that they have trouble with some of the more unusual autosign templates, and was hoping you would be able to fix them by adding exceptions for each template after the (UTC) portion of the timestamp (noting that they should all be substituted... I guess that's how you've done it for Template:Unsigned), ideally with an additional fixed text afterwards in the case of Template:Unsigned2Fix and Template:UnsignedIP2Fix. For an example of the transcluded/substituted Template:Undated not working see the top section of Wikipedia talk:WikiProject Novels.
I've also proposed that the templates be edited to a consistent style here, so if you have the time and could let us know what's helpful for archiving purposes, please leave a comment there.
Many thanks. --xensyriaT 11:42, 28 May 2012 (UTC)[reply]
- Here's an example :) --xensyriaT —Preceding undated comment added 14:03, 11 June 2012 (UTC)[reply]
I am having trouble with the Auto Archive on my talk page, I believe I have it set up right, but it has been a couple of days and the bot still hasn't archived my talk page, can you take a look at it and see if I set something up wrong? Thanks --Clarkcj12 (talk) 18:51, 31 May 2012 (UTC)[reply]
- I moved the closing "}}" to its own line. Maybe that was the problem.
- —WWoods (talk) 20:57, 31 May 2012 (UTC)[reply]
- Thanks for trying User:Wwoods but it still won't work. --Clarkcj12 (talk) 21:16, 31 May 2012 (UTC)[reply]
- The bot hasn't worked its way around to you yet: User contributions:MiszaBot III. Wait, and see what happens circa 14:00 (UTC).
- —WWoods (talk) 04:20, 1 June 2012 (UTC)[reply]
It seems to be struggling to correctly archive my talk page. If you take a look at the last two or three times it edited my talk page, it seems to be struggling with adding material to the two existing archive pages. It also seems to be incapable of simply creating another archive page - it "complains" that the existing pages are full. It is also completely ignoring the oldest content on my talk page. I'd prefer to have the archive content be in the correct chronolgical order. Roger (talk) 18:18, 3 June 2012 (UTC)[reply]
- I made these changes. MiszaBot is sometimes picky about whitespace and the order of the parameters. The bot wasn't changing its own code to reflect that it should be in archive 2; it was stuck on archive 1 for some reason; I fixed that too. Keep in mind that the bot will not archive sections that do not have timestamps; you have many sections without timestamps. Brad (talk) 20:07, 3 June 2012 (UTC)[reply]
- Thanks. I'll have to look up the page history then add correct timestamps to the oldest posts.
- If you add timestamps to match the date the sections were created the bot will not sort them chronologically except to place them in the current archive. If you want them sorted exactly when they were made you'll have to do it manually. Brad (talk) 20:43, 3 June 2012 (UTC)[reply]
- I noticed that the bot appears to have been archiving to both Archive 1 and Archive 2 so there may be a small knot to untangle. I don't know why. Also, what happened to all of the archives that are seen in this edit summary? Sometimes, other archive names cause conflicts.
⋙–Berean–Hunter—► 20:54, 3 June 2012 (UTC)[reply]
- Well, they're out there: [1]. Merging them all back.... —WWoods (talk) 00:23, 4 June 2012 (UTC)[reply]
- Why not wait for Roger to tell us what is going on before you start merging. Brad (talk) 00:50, 4 June 2012 (UTC)[reply]
- Oh! No, I was just thinking out loud about the size of the task, not announcing that I was doing it.
- —WWoods (talk) 19:36, 5 June 2012 (UTC)[reply]
- This page still hasn't archived and I believe it is due to the other quasi-random archive page titles which will probably need straightening out. Merging back may be inevitable here.
⋙–Berean–Hunter—► 13:16, 8 June 2012 (UTC)[reply]
- The bot settings are at 100d in age and there isn't anything on the page which has passed that mark yet although some are very close. The rest of them are lacking timestamps. Brad (talk) 13:32, 8 June 2012 (UTC)[reply]
- I see now. I quickly scanned and saw the "Ibid" thread dated 2009 and didn't look further. Cheers,
⋙–Berean–Hunter—► 14:16, 8 June 2012 (UTC)[reply]
MiszaBot at Commons bot stopped after another /dev/null archiving and I just stumbled about a similar behaviour here. It does seem to be a problem how the MiszaBot/config template is placed at the talk page as archiving worked fine on this page until the bot template was modified into a new (incorrectly formatted?) format by user Yutsi. --Denniss (talk) 21:45, 5 June 2012 (UTC)[reply]
- The code needs to be on separate lines Brad (talk) 22:25, 5 June 2012 (UTC)[reply]
- That's what I was assuming. But the bot needs a failsafe function to prevent removal of discussions if it isn't able to identify the archive location. --Denniss (talk) 09:54, 10 June 2012 (UTC)[reply]
MiszaBot has stopped working for the archiving of Talk:Mitt Romney dog incident since the name of the page was modified. The name of the old archives was updated, but no new archiving has occurred. NJ Wine (talk) 13:06, 8 June 2012 (UTC)[reply]
- I've made a few tweaks but nothing that would be an obvious fix.
⋙–Berean–Hunter—► 13:46, 8 June 2012 (UTC)[reply]
- Thanks. NJ Wine (talk) 17:12, 8 June 2012 (UTC)[reply]
The bot was working as programmed. Unless |minthreadsleft = and |minthreadstoarchive = are used the defaults are 5 and 2 respectively. In other words the bot will not archive anything until the page has 7 sections and once it does the bot will archive 2 of them leaving 5 on the page. Brad (talk) 04:04, 9 June 2012 (UTC)[reply]
Anyone care to take a look at Talk:X-Men: First Class and see why the automatic archiving is not working? Cheers. Betty Logan (talk) 20:27, 9 June 2012 (UTC)[reply]
- Hard to say; the settings looked right. I did a manual archiving and moved the bot code to the top of the page. See what happens. Brad (talk) 22:32, 9 June 2012 (UTC)[reply]
MiszaBot I (talk · contribs) is not completing a full cycle through all of the talk pages. The last two runs it stopped in the I's and the M's. . Brad (talk) 07:32, 11 June 2012 (UTC)[reply]
MiszaBot II (talk · contribs) Since June 6th it has only been archiving the administrators noticeboard. Brad (talk) 07:07, 11 June 2012 (UTC)[reply]
MiszaBot III (talk · contribs) has not run since June 8th.
This is definitely the reason people are reporting no archiving. Brad (talk) 07:32, 11 June 2012 (UTC)[reply]
- MiszaBots II & III appear to be working correctly now.
⋙–Berean–Hunter—► 15:51, 11 June 2012 (UTC)[reply]
- MiszaBot II has still not visited a great many pages that it normally does. One example, Wikipedia talk:Good article nominations, where the last archive was on May 25, and the next could have been made as early as June 1, but has yet to be picked up. BlueMoonset (talk) 18:48, 11 June 2012 (UTC)[reply]
- Progress report on MiszaBot III: it made it to "M" today before stopping (it ran 12:35 through 20:54). On June 8, it stopped during "P" (12:37 through 18:45). MiszaBot II ran for about an hour this morning (09:30 through 10:32) before stopping, and then did a single pickup from the Administrator's noticeboard at 14:38. BlueMoonset (talk) 22:10, 11 June 2012 (UTC)[reply]
- June 12:
- MiszaBot II has only done every-eight-hour pickups from the Admin noticeboard.
- MiszaBot III stopped at "Penyulap" today after running 12:41 through 20:34. It hasn't done a complete archive since June 6, and stopped after Penyulap on June 8. BlueMoonset (talk) 23:24, 12 June 2012 (UTC)[reply]
The same problems with MiszaBot at Commons, very intermittent runs, stops midway through an archive process (saves to archive but doesn't remove from talk page). Is there a problem with the Bots or is it caused by lags in the Wiki infrastructure (toolserver, database server) ? --Denniss (talk) 00:08, 13 June 2012 (UTC)[reply]
- I'm not sure what's the problem yet; the bots start their runs, and toolserver seems responsive enough. However, looking into logs, I see that at some point it just receives a timeout from en.wikipedia.org and falls into an infinite loop of (failed) retries. —Миша13 07:00, 13 June 2012 (UTC)[reply]
- The bot was archiving WP:ANI and the past few archive operations it sent threads to the archive but did not remove them from the main talk page. Brad (talk) 19:41, 13 June 2012 (UTC)[reply]
- Also chiming in from WP:DRN where I just moved 76k of text that was targeted for Archiving. Hasteur (talk) 20:39, 13 June 2012 (UTC)[reply]
- Any sort time frame on fixing this? Andrewcrawford (talk - contrib) 09:54, 19 June 2012 (UTC)[reply]
The bots are bypassing various pages and it seems to be random. WP:SHIPS was one of them but I also notice that this very talk page hasn't archived either. I've converted several pages to ClueBot III. Brad (talk) 00:49, 28 June 2012 (UTC)[reply]
Miszabot seems to have functioned properly, despite not having each parameter on a separate line
(Talk:Sicilian_Defence, &
/Archive_1).
Was a change made so that's no longer required? Or did I misunderstand the requirement?
—WWoods (talk) 23:08, 14 June 2012 (UTC)[reply]
- First of all, it seems that many of the recent problems were caused by my pywikipedia checkout going out-of-date. My nightly
svn update job has been (unknowingly to me) failing for the last few months. After I fixed it manually, it seems to have started running a bit more smoothly. Other thing I noticed, user:xqt has been "sneaking in" fixes into my bot's code during that time (many thanks!). One of those is a new configuration parser, which allows the bot to finally read it properly even if it is squished into one line. —Миша13 08:59, 15 June 2012 (UTC)[reply]
Hi Misza13,
(long post follows... sry)
I'd like to set up automatic archiving for the Guest intro boxes on Teahouse. Currently, we're transcluding two sub-pages of Teahouse/Guests as a (clever?) layout hack, using one subpage for the right column and one for the left column. Each new intro is added as a new section to one of these two pages. This allows us to keep the process of introducing yourself relatively easy for new editors, while keeping the resulting page sleek and pretty. Feel free to try it yourself with a test introduction, to see what I mean.
We want to archive these pages automatically, to keep the # of guest intros displayed on the /Guests page to a reasonable number, and to only display the most recent guests there. 'Older' intros are currently being moved (manually) to the Guestbook. We'd like to turn the Guestbook into a paginated and searchable archive, to keep it from getting too long and to facilitate browsing.
I've recently added an automatic timestamp to all newly-created guest intros, which should (if I'm reading the pywikibot code right) make it possible for them to be automatically archived by Miszabot II. You can see the first example of this timestamped intro here: Wikipedia:Teahouse/Guests/Left_column#Capt_Swing.
I would like to set things up so that from now on all intros from both the /Right_column and /Left_column pages are archived to the Guestbook, rather than creating two separate archives. Is this possible with Miszabot? Reading through your (helpful) how-to doc, it looks like I can do this if you give me a unique key value for the key parameter. So I guess my questions, finally are:
- do you think what I'm describing is doable with MiszaBot, and
- if so, are you willing to give me the necessary key value?
Cheers, - J-Mo Talk to Me Email Me 23:35, 14 June 2012 (UTC)[reply]
I don't know whether it's appropriate to post on a user talk page just to say hello, but that's all I'm doing. :) I got to your user page through a link on Riley Huntley's page. He had posted something on my old IP address talk page and I was following up to let him know that I now have a named account. Anyway, I love the way your user page is set up and the way it looks. And by the way, your reference to /dev/null above makes it sound like you're a fellow Linux user. Glad to meet you. Guyovski (talk) 01:09, 18 June 2012 (UTC)[reply]
Hi,
The Talk pages are greatly harmed by questions being archived when they're still waiting for a response. Sometimes a question has to sit on a talk page for 2, 3, or 5 years before the right person notices it and explains the issue.
I was trying to think of a good policy for setting the time (2 years seems good, but sometimes that's too short), then I realised that they shouldn't be archived at all based on time.
Why are they ever archived? I guess because the pages become enormous and threads start getting lost and repeated. So then the criteria should be based on how big the pages is (kb) or how many threads are there. Time shouldn't be a factor at all.
Can this be done? What do you think? Gronky (talk) 18:06, 19 June 2012 (UTC)[reply]
- ok lets say we do it your way, what abouta spammer, or none spammer, someone who maybe creates new talk discussion but never get a reply as you say, but if oyu go on count it would do th first one or even size, ther eno way to determine which to archive, unless the bot is made inteligent so it can read and deicded if the talk is require it probally never work in any format that really be useful for anyone, the bot probally can be made to do something what you saying but it wont be perfect that way is the point i am makingAndrewcrawford (talk - contrib) 18:24, 19 June 2012 (UTC)[reply]
- Spam is a separate problem. Archiving pages isn't an anti-spam measure. If there's spam, it should be removed, not archived. Archive bots don't have to detect spam or understand the content of pages. If there is a situation where the bot doesn't know what to do, then the right thing to do is do nothing. Wikipedia is written by thousands of humans, so when intelligence is needed, let the humans do it. Bots should only do simple, boring, repetitive work.
- No behaviour will ever be perfect, but the current behaviour has a problem and I've proposed a fix. In light of my comment about the content of pages being not important, do you now think it's possible? Gronky (talk) 20:09, 19 June 2012 (UTC)[reply]
- Hi. There are a number of options available for the bot to consider before archiving. For example, the bot accepts minthreadstoarchive and minthreadsleft parameters, as described in the documentation.
Ideally, LiquidThreads or some variant will replace the entire current talk page system. Don't hold your breath, though. In this particular case, it sounds as though pages are simply being set archived by bot when that isn't necessary. The answer there is to disable the archiving by bot. Archive bots are generally intended for (very) active talk pages; they're certainly not run over every talk page and there are thousands of talk pages with threads from two, three, four, or even five years ago. There are a few other archive bots, one or two of which may support only archiving threads marked with a particular template (e.g., {{resolved}}). Again, though, the real answer here is to stop using talk pages in this manner and rely on a proper communication system such as LiquidThreads or the (allegedly upcoming) Echo framework. --MZMcBride (talk) 20:32, 19 June 2012 (UTC)[reply]
- Thanks for this reply.
- I agree that the problem is mostly caused by incorrect use of the bot(s), but it's still worth thinking about how the bot could be changed to make incorrect use less likely. Counting in days was probably a bad idea from the start. Weeks or months would have been a better. (This would also avoid the abuse I've heard of where people set the bot to 3 days as a way to disrupt discussion and hide previous discussion.) The documentation could mark "days" as deprecated, a bot could convert all uses of days to the equivalent in weeks (round up), and "days" could be removed.
- Or for low values (< 6 months), a click by a human could be necessary.
- These problems can be fixed or reduced if the will is there. Gronky (talk) 23:04, 19 June 2012 (UTC)[reply]
- Are you volunteering? :-) --MZMcBride (talk) 05:28, 20 June 2012 (UTC)[reply]
Talk:Acid house/Archive 7, a page you substantially contributed to, has been nominated for deletion. Your opinions on the matter are welcome; please participate in the discussion by adding your comments at Wikipedia:Miscellany for deletion/Talk:Acid house/Archive 7 and please be sure to sign your comments with four tildes (~~~~). You are free to edit the content of Talk:Acid house/Archive 7 during the discussion but should not remove the miscellany for deletion template from the top of the page; such a removal will not end the deletion discussion. Thank you. SabreBD (talk) 07:25, 25 June 2012 (UTC)[reply]
Thought you'd like to know about this. The bot didn't grab all the text of a section to be archived at Wikipedia talk:Manual of Style/Layout. This section, "Section templates and summary style," was to be archived, but after the bot snipped only part of the text, it looked like this due to a now unclosed pre tag enveloping the rest of the talk page. I fixed everything manually, but figured you'd like to know if there is some reason the bot is identifying pre tags as the end of a section. Cheers, Rkitko (talk) 12:36, 26 June 2012 (UTC)[reply]
Hi! Is it possible to share the code of your bot with me or possibly put it into open source? I'm an administrator of the Ukrainian Wikipedia and would like to set up autoarchiving there. Thank you in advance! --DixonD (talk) 19:56, 29 June 2012 (UTC)[reply]
- I second this: if you're willing to share it would be very useful for trying to fix up the autosign templates. Thanks :) --xensyriaT 17:09, 30 June 2012 (UTC)[reply]
|
